Human Computer Interaction
AI-Powered Playbook for Enhancing Focus Group Methodologies
This paper proposes a playbook for integrating AI into focus groups, leveraging AI's capabilities in live conversation scaffolding. The playbook organizes AI supports by role (tool, co-host, host) and modality (text, voice, embodied), addressing the need for structured application of AI in User Expe…
Novice Users' Struggles with AI-Generated Visualizations Uncover Risks and Model-Specific Failures
This study investigates the challenges faced by visualization novices when using conversational AI (ChatGPT, Gemini, Claude) to generate and interpret data visualizations. A user study with 20 novices revealed issues related to AI execution compliance, visualization quality, user misuse, and factors…
Co-Designing Accessible Multi-Modal 3D Data Visualization for Blind Users Without Visual Dependency
Researchers used Experience-Based Co-Design with five BLV experts and one sighted researcher to develop a web-native, non-visual 3D visualization tool featuring sonification, stereo/volumetric audio, and buffer aggregation. The multi-phase process compared tactile and digital probes, targeting analy…
Speech Recognition Outperforms Touch Keyboards by Nearly 3x for Short Messages in English and Mandarin
Laboratory tests on iPhone 6 Plus show Baidu Deep Speech 2 achieving 153 WPM in English and 123 WPM in Mandarin for short message transcription, versus 52 WPM and 43 WPM for iOS Qwerty and Pinyin keyboards. Speech input rates are 2.93x faster in English and 2.87x faster in Mandarin under ideal condi…